Dynamics of Fishers' Responses to Social-Ecological Change in Coastal Mozambique: A Resilience Perspective

Dynamics of Fishers’ Responses to Social-Ecological Change in Coastal Mozambique: A Resilience Perspective

By

Jessica Blythe

B.Sc., Memorial University, 2004 M.A., York University, 2009

A Dissertation Submitted in Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY

in the Department of Geography

© Jessica Blythe, 2013 University of Victoria

All rights reserved. This dissertation may not be reproduced in whole or in part, by photocopy or other means, without the permission of the author.

Dynamics of Fishers’ Responses to Social-Ecological Change in Coastal Mozambique: A Resilience Perspective

By

Jessica Blythe MA, York University, 2009 BSc, Memorial University, 2004

Supervisory Committee:

Dr. Mark Flaherty, Co-Supervisor

(Department of Geography, University of Victoria)

Dr. Grant Murray, Co-Supervisor

(Department of Geography, University of Victoria)

Dr. Steve Cross, Department Member

(Department of Geography, University of Victoria)

Dr. Jack Littlepage, Additional Member

Supervisory Committee:

Dr. Mark Flaherty, Co-Supervisor

(Department of Geography, University of Victoria)

Dr. Grant Murray, Co-Supervisor

(Department of Geography, University of Victoria)

Dr. Steve Cross, Department Member

(Department of Geography, University of Victoria)

Dr. Jack Littlepage, Additional Member

(Centre for Global Studies, University of Victoria)

Abstract

Change has become a ubiquitous force in a highly globalized and interconnected world. Coastal systems are being restructured by overfishing, globalization, climate change and other factors. Further, social and ecological changes in coastal systems interact across spatial and temporal scales creating challenges that are complex, nonlinear and often difficult to predict. These new challenges have the potential to push social-ecological systems past their experienced range of variability and thus have immense consequences for the both the health of marine ecosystems and the livelihoods of hundreds of millions of people who depend on marine resources for their survival. Addressing these challenges will require collaborative efforts informed by site-specific research on the dynamics of social-ecological systems.

Ultimately, this dissertation aims to contribute to efforts towards social-ecological system sustainability. Specifically, the purpose of the research is to improve our understanding of how small-scale fishers in Mozambique have adapted over time to cope with a particular set of challenges and how likely fishers are to cope effectively with future changes in their complex social-ecological systems. The dissertation is organized around four research chapters, each of which addresses a specific research objective.

Detailed knowledge of historical social-ecological conditions is a critical entry point for

understanding small-scale fisheries systems. While fisheries landings data are often the primary source for historical reconstructions of fisheries, reliance on data of a single type and/or from a single-scale can lead to incomplete or misleading conclusions. Moreover, in the case of many small-scale fisheries landings statistics are often incomplete and/or inaccurate. Therefore, Chapter 2 combines data from multiple sources and scales to reconstruct historical social-ecological system dynamics along the Mozambican coast. At the national scale, my analysis points towards trends of fishing intensification and decline in targeted species, and highlights the significant impact of small-scale fisheries on marine stocks. At the local scale, fishers are

experiencing changes in fish abundance and distribution, as well as in their physical, social and cultural environments and have responded by increasing their fishing effort.

In the context of multiple drivers of change, it has become increasingly important to identify how communities are responding to livelihood stressors. In Chapter 3, I examine how fishers are adapting to social-ecological change, and identify factors that facilitate adaptation and factors that inhibit adaptation. Primarily, fishers are adapting through intensifying their fishing efforts or by diversifying their livelihoods. Adaptation is facilitated by fishers’ groups, occupational pride and family networks. It is inhibited by limited assets, adaptive actions with negative social and ecological impacts, competition over declining resources and pervasive poverty. My data suggest that it is not the poorest fishers who are least able to adapt to change, but fishers who are locked into a declining fishery. I argue that adaptations are spatially and social differentiated and place-specific. Therefore, future adaptation initiatives aimed at strengthening the capacity of threatened communities to respond to livelihood stressors need to explicitly consider this complexity.

Millions of people around the world depend on shrimp aquaculture for their income and livelihood. Yet, the phenomenal growth of shrimp aquaculture during the last two decades has given rise to considerable environmental damage and social disruption at the local level. In Chapter 4, I analyze the impacts of employment at an export-oriented shrimp farm in central Mozambique on livelihood vulnerability of farm and non-farm employees. My data indicate that shrimp farm employees are less vulnerable to chronic stressors, such a pervasive poverty, than

non-farm employees, but more vulnerable to acute shocks, such as the White Spot Syndrome Virus (WSSV), associated with shrimp production than non-farm employees. I argue that future vulnerability research will need to account for this duality as aquaculture development spreads along the Western Indian Ocean.

In response to the speed and magnitude of contemporary change, understanding how much disturbance communities will absorb, where social thresholds lie and what coastal community systems might look like after a threshold is crossed are critical research questions. Chapter 5 evaluates the resilience of two fishing communities in central Mozambique and forecasts the outcome of moving past socially defined thresholds. My results indicate that coastal

communities are continuously absorbing multiple sources of disturbance without shifting into different states. However, a 90% decline in catch rates would represent a threshold for both communities. At Zalala Beach, fishers would respond by permanently moving to another location whereas in Inhangome, fishers would respond by changing their professions. These results contribute to our understanding of social resilience.

Deliberate progress towards the goal of long-term sustainability depends on understanding the dynamics of social-ecological systems. Therefore, this dissertation aims to contribute to a growing body of theory and empirical evidence on how fishers negotiate livelihoods under conditions of rapid change and increasing vulnerability. The dissertation concludes by summarizing seven key research findings and by discussion some of the theoretical, methodological and policy contributions of my research to the literature.

Acknowledgments

This dissertation is a collaborative project that would not have been possible without the support of many valued friends and colleagues. Innumerable people provided me with the assistance I needed to finish this project. To all of them, I am indebted.

First and foremost, I owe my dissertation to the fishers and families along the central coast of Mozambique who taught me about their history, culture, spiritual beliefs, fishing practices and ecological knowledge and shared with me their hopes and fears. Without their generosity, patience and knowledge, my research would not have been possible.

I owe many thanks to my supervisors. I am grateful to Dr. Mark Flaherty for bringing me on board with the Southern Ocean Education and Development (SOED) project, which allowed me to step into the world of fishing in Mozambique. I would like to thank Dr. Grant Murray for his mentorship, intellectual guidance, network connections and meaningful contributions throughout the writing process. Thank you to Dr. Jack Littlepage and Dr. Steve Cross, members of my academic committee, for their support throughout my PhD program. Thank you to Dr. Melissa Marschke for agreeing to be the external member of my academic committee and for her insightful comments on this dissertation.

To my Mozambican family and colleagues, I am most grateful. Ana Cristina, in 2009 you warmly opened your house to me which became my home for the remainder of my fieldwork, you picked glass out of my hair after our accident in 2010 and you welcomed my family into yours in 2012. I will be forever grateful for your friendship and look forward to the next time we walk along the coast together. Graham, Jenny and Sana, thank you for sharing your home, wife and mother with me! Valdemiro, without your skills, passion and insights in the field, this work would not have been possible nor would I have understood Mozambique as well as I do today. Erica, Naftal, Jeremias, Pita, Dona Rosa and everyone I worked with at UEM-ESCMC thank you for your friendship and support throughout my work in Mozambique. Francois, thank you for your time, generous support and our discussions over coffee. Honorio, thank you so much for your help at Zalala Beach. Finally, Dr. Hoguane, thank you for welcoming into UEM-ESCMC.

Thank you to SSHRC, John Montalbano and the Centre for International Governance

Innovation’s African Initiative, who funded this work and ensured that I spent adequate time spent in the field, analyzing and writing my dissertation.

A huge thank you to my team here at UVic. Regina, you saved me more times than I can remember. Thank you so much for taking care of me while I was overseas, for helping with translations and for much needed coffee breaks! Laurie, working at the Writing Centre was an incredibly formative time for me. Thank you for your mentorship. Darlene, thank you for always making sure I was staying on track. Dr. Phil Dearden, thank you for your mentorship and support throughout my program.

Next, to my dear friends. To Kylee and Erin, DANCE lab forever! Thank you for the hundreds (thousands?) of hours spent talking together about research and about life in general. Thank you Maral and Jenna, my colleagues in writing. Kat Middleton, you’ve had my back from the very beginning. Word. And to all the family and friends who helped keep me sane along this journey, thank you.

Mom and Dad, thank you for loving me, supporting me, coming to Mozambique with me and always being proud of me. Dad, you tackled chemo with courage, optimism, curiosity, humour and a rigorous scientific method! Your approach to your cancer inspired me throughout my dissertation and will continue to inspire me as I take on life’s next adventure. And lastly, to my husband Phil, who makes everything more fun! Thank you for moving to Victoria with me so I could follow this dream and for your unwavering love and support.

Table of Contents

Supervisory Committee Page

Abstract

Acknowledgements

Table of contents

List of tables

List of figures

Acronyms

Glossary

Chapter 1: Introduction

Research context and objectives………...…3

Research objectives……….………..……...4 Analytical framework………..……4 Resilience………..……..5 Social-ecological systems………..…….5 Historical contexts………..…6 Adaptation………..…6 Vulnerability………..…….7 Fishers’ knowledge……….…....8 Livelihoods………...……..…8

Introducing the case study………..……….9

Why Mozambique? ……….………..…….9

The small-scale fishery in Mozambique………...…11

Research approach………....11

Place-based, case study analysis………..………11

Mixed methods……….………….12

Field research……….…..12

Chapter 2: Historical perspectives and recent trends in the coastal

Mozambican fishery

Results and discussion………..………….22

Macro-scale restructuring in the Mozambican fisheries: 1950-2009………..………22

Micro-scale restructuring: fishers’ knowledge………25

Conclusion……….31

Chapter 3: Strengthening threatened communities through adaptation:

insights from coastal Mozambique

Abstract……….………35

Introduction………...…36

Drivers of change in the small-scale fishery……….………38

Methods……….54

Research communities………..39

Data collection………...………..39

Results and analysis………..42

Fishers’ assets………..42

Adaptive responses to livelihood stressors………...………44

Adaptive strategies: intensify or diversify………45

Factors that facilitate adaptation……….……46

Factors that inhibit adaptation……….……47

Conclusion……….49

Chapter 4: Impacts of export-oriented shrimp aquaculture on livelihood

vulnerability: a case from the Western Indian Ocean

Abstract……….…53

Introduction………...……54

Vulnerability framework………...…56

Brief history of aquaculture in Mozambique………59

Methods……….…61

Study site………..….61

Surveys and interviews……….……63 Results………..……….……64 Exposure………..………….……64 Sensitivity………..………66 Adaptive capacity……….……67 Synthesis……….……….….70 Discussion……….72 Conclusion……….……77

Chapter 5:

Assessing the sustainability of fishing communities: insights from

resilience thinking………...………79

Abstract……….80

Introduction………...………81

Resilience, thresholds and surrogates……….……..82

Methods………..…….…..85

Data collection………...……..…85

Study sites………..………...87

Results and analysis………..…88

Step one – system identity………..……...88

Step two – thresholds………..…..…92

Step three – future scenarios………..…..…96

Conclusion………..…...…98

Chapter 6: Conclusion

Principle research findings……….……….101

Limitations of the research……….……….106

Contributions to the literature………...………..107

Contributions to theory………..107

Contributions to methods……….………..109

Contributions to policy……….………..110

Literature Cited

Appendix A: Historical interviews

Appendix B: Household surveys

Appendix D: Aquaculture interviews

List of Tables

Table 1.1 Timeline of fieldwork in Mozambique……….………13

Table 2.1 Common observations of change as described by fishers at Zalala Beach…………...26

Table 2.2 Average reported daily catch (kg) per lancha boat at Zalala Beach……….……27

Table 3.1 Summary of assets across the two study communities………...…..43

Table 3.2 Adaptive strategies in response to livelihood stress across the study areas…….…….45

Table 4.1 Summary of assets in Inhanssunge, Mozambique………...….67

Table 4.2 Adaptive strategies used in response to stressors in Inhanssunge, Mozambique….…69

Table 4.3 Vulnerability analysis for shrimp farm employees and non-farm employees in

Inhanssunge, Mozambique………71

Table 5.1 Elements of the Sofala Bank social-ecological system’s identity, as identified in

List of Figures

Figure 1.1 Small-scale fishers in Mozambique……….………..…2

Figure 2.1 Map of Mozambique………...………20

Figure 2.2 a) Periods of socio-economic change affecting coastal fisheries and b) total landings and total number of small-scale fishers in Mozambique………..……….22

Figure 2.3 A lancha boat and its smaller predecessor, a traditional dugout canoe at Zalala

Beach……….………30

Figure 3.1 Location of the research communities, Inhangome and Zalala Beach, in

Mozambique………..……40

Figure 3.2 Characteristic multiple-species catch by small-scale fishers in central

Mozambique………..……41

Figure 4.1 Vulnerability framework……….……58

Figure 4.2 Location of the shrimp farm in Inhanssunge district, central Mozambique……...….62

Figure 4.3 Satellite image of the shrimp farm in Inhanssunge Mozambique………...63

Figure 4.4 Summary of exposure to stressors and shocks in Inhanssunge, Mozambique………64

Figure 5.1 ‘Ball in a basin’ heuristic depicting multiple stability domains for social-ecological systems………...…83

Figure 5.2 Focus group with small-scale fishers at Zalala Beach………...…87

Figure 5.3 Location of the research communities, Inhangome and Zalala Beach, along the Sofala Bank, Mozambique………..…..…88

Figure 5.4 Fishers’ responses to hypothetical declines in catch rates from two communities in central Mozambique………..…….94

Acronyms

CCP Conselho Comunitário de Pesca (community fishing council) FAO Food and Agricultural Organization of the United Nations

FRELIMO Frente de Libertação de Moçambique (Front for the Liberation of

Mozambique)

IDPPE Instituto de Desenvolvimento de Pesca de Pequena Escala (Institute for the Development of Small-Scale Fisheries, Government of Mozambique) IIP Instituto de Investigação Pesqueira (Institute for Fisheries Research,

Government of Mozambique) IMF International Monetary Fund

IPCC Intergovernmental Panel on Climate Change MZN Mozambican metical (national currency)

NACA Network of Aquaculture Centres in Asia-Pacific

PCR Poupança e Crédito Rotativo (saving and lending groups)

PI Principle Investigator

ProPESCA Artisanal Fisheries Promotion Project funded by the International Fund for Agricultural Development

RENAMO Resistência Nacional Moçambicana (Mozambique National Resistance) UEM-ESCMC Universidade Edaurdo Mondlane – Escola Superior de Ciências Marinhas

e Costeiras (University of Eduardo Mondlane - School of Marine and

Coastal Sciences)

UNDP United Nations Development Programme UNEP United Nations Environment Programme

WB World Bank

WSSV White Spot Syndrome Virus

WCED World Commission on Environment and Development WWF World Wildlife Fund

Glossary

Canoa tipo moma Wooden fishing vessel (9-10 meters in length)

Chuabo Local language, commonly spoken in central Mozambique

Combinados pesqueiro Fishing cooperative

Comprador Fish trader or middlemen

Courandeiro Traditional healer

Lancha Wooden fishing vessel (9-10 meters in length)

Machamba Subsistence garden

Régulo Traditional leader

Secretaria do Bairro Village head

Research Context and Objectives

He was an old man who fished alone in a skiff … and he had gone eighty-four days now without taking a fish.

- Hemmingway (1952)

The importance of marine resources for global livelihoods cannot be overstated. Hundreds of millions of people worldwide are directly dependent upon fishing for their survival (Allison and Ellis 2001, Allison et al. 2009). This dependence on fisheries is particularly acute for poor and marginalized populations in developing countries (Béné et al. 2007). As Hemmingway alludes in the above quotation, fishers are accustomed to adapting to variability within an unpredictable resource base. Yet, Hemmingway wrote the story of Santiago, an aging Cuban fisher and his relationship with the sea, years before industrial fishing, globalization and climate change challenged the resilience of even the most robust fishers. Today, after more than six decades of systematic exploitation, global fisheries are in crisis (Pauly et al. 1998, 2002). The ocean’s capacity to provide food, maintain water quality and recover from periods of stress is being eroded by overfishing (Worm et al. 2006). Multiple factors, including sand mining, tourism, oil and gas and the internationalization of the shrimp trade, are driving coastal habitat loss, making people more vulnerable to coastal disasters (Adger et al. 2005b, Primavera 2006). These new challenges have the potential to push social-ecological systems past their experienced range of variability and thus have immense consequences for the health of marine ecosystems and the well-being of people whose livelihoods depend on fisheries (Perry et al. 2010). Addressing these challenges will require collaborative efforts informed by site-specific research on the

vulnerability of fishers and their ability to cope with and adapt to social-ecological change.

Ultimately, this dissertation aims to contribute to the efforts towards social-ecological system sustainability, that is, the use of natural resources to meet the need of the present without compromising the ability of future generations to meet their own. Specifically, the purpose of my research is to improve our understanding of how small-scale fishers have adapted over time to cope with a particular set of challenges and how likely they are to cope effectively with future changes in their complex social-ecological systems. I focus on resilience and livelihood

dynamics to investigate how and why some people, places or groups are adaptive while others shift into undesirable states.

Research Objectives

Deliberate progress towards the goal of long-term sustainability depends on understanding the dynamics of social-ecological systems (Cumming et al. 2005). Many fisheries studies have examined some aspect of human-environment interactions in coastal systems, however the complexity of coupled social-ecological systems over time has not been well understood (Mahon et al. 2008). The dissertation aims to contribute to a growing body of theory and empirical evidence on how fishers negotiate livelihoods under conditions of rapid change and increasing vulnerability. Specifically, this dissertation has four objectives:

1. To reconstruct historical social-ecological system dynamics by combining data from multiple sources and multiple scales.

2. To examine how fishers are adapting to social-ecological change, and to identify some of the key factors that facilitate adaptation and some key factors that inhibit it.

3. To analyze the impacts of employment at an export-oriented shrimp farm on livelihood vulnerability.

4. To assess the resilience of livelihood systems and forecast the outcome of moving past socially defined thresholds.

I argue that by learning as much as possible about the dynamics of fishers’ responses to change, research can provide insights which have the potential to improve fisheries policy and

governance and contribute towards social-ecological sustainability.

Analytical Framework

To address these research questions, I situate the experiences of small-scale fishers within the literature on resilience, adaptation, vulnerability and livelihoods. This section introduces the central theoretical approaches used throughout the dissertation. Further development of each concept is integrated throughout the remainder of the chapters. For example, in Chapter 3 I combine adaptation and livelihood literature to explore how adaptation is facilitated, inhibited and socially differentiated at the household and community scale. In Chapter 4, I probe the

boundaries of the vulnerability literature by proposing that a single factor, namely employment at a shrimp farm, can simultaneously increase and decrease social vulnerability. In Chapter 5, I explore system resilience and the utility of socially defined thresholds for sustainability analysis.

Resilience

The scale and complexity of contemporary social and ecological change, combined with the failure of our traditional governance approaches to successfully manage natural resources, motivated scholars from a range of academic disciplines to look for new ways to understand the relationships between social and natural systems. One of the ideas to emerge from these

interdisciplinary efforts is the concept of resilience. In 1973, C.S. Holling published a seminal paper on resilience and stability in predator-prey systems. Counter to the dominant thinking at the time, which emphasized equilibrium or the balance of nature, Holling suggested that systems could persist in multiple stable states. He argued that “resilience determines the persistence of relationships within a system and is a measure of the ability of these systems to absorb changes of state variables, driving variables, and parameters, and still persist” (Holling 1973, p. 17 as cited in Folke 2006). The concept was quickly adopted by academics in anthropology,

environmental psychology, human geography, common property research, economics and other disciplines (Folke 2006). Today, resilience is defined as the capacity of a social-ecological system to tolerate disturbance without collapsing into a qualitatively different state that is

controlled by a different set of processes (Gunderson and Holling 2002, Walker et al. 2004). My research approaches small-scale fishers’ responses to change through the lens of resilience theory.

Social-Ecological Systems

One of the central pillars of resilience thinking is that human and natural systems are not conceived as separate systems. Rather, resilience thinkers argue that these two components are indivisibly linked and that systems should be seen as social-ecological systems (Berkes and Folke 1998).1 The term social-ecological system is used to emphasize that the two components are equally important, that they function as a coupled, interdependent and interactive system and

1 _{Also called coupled human-environmental systems (Turner et al. 2003) and socio-ecological systems (Gallopin} 2006).

to stress that the delineation between subsystems is artificial (Berkes et al. 2003). In social-ecological systems thinking, it is understood that a two-way feedback relationship exists between system subcomponents: human actions affect biophysical systems, biophysical factors affect human well-being and humans in turn respond to these factors (Ommer et al. 2011). Social-ecological systems thinkers also stress that knowing the history of social-Social-ecological

transformations over time is critical to understanding the contemporary context within which individuals and communities respond to change (Ommer 2007). The need to investigate the whole social-ecological system represents an important departure from traditional fisheries science research which largely treated the study of biophysical systems as separate from the study of human subsystems (Berkes 2011). The analysis of fisheries livelihoods in this dissertation is made in the context of social-ecological systems.

Historical Contexts

Fundamentally within resilience analyses, social-ecological systems are understood to be the product of centuries of interactive restructuring between people and natural environments

(Ommer 2007). Studies of social-ecological systems must, then, begin with thorough analysis of the history of social-ecological interaction within the system of interest (Walker et al. 2002, Seixas and Berkes 2003). Yet, many social-ecological systems analyses tend to be static. As Scoones observes, many resilience studies have “tended to ignore questions of dynamics and variability across time and space, often excluding from the analysis the key themes of

uncertainty, dynamics and history” (1999, p. 480 as cited in Ommer 2007). This dissertation responds to calls for more nuanced understandings of the complex historical interactions between fishers and marine ecosystems by beginning with a historical analysis of the coastal fishery in Mozambique (Murray et al. 2006, St. Martin et al. 2007). As Berkes et al. (2003, p. 8) outline, “it is difficult or impossible to understand a system without considering its history, as well as its social and political contexts”.

Adaptation

Researchers have generally approached studies of adaptation from two distinct traditions: global environmental change and resilience (Nelson et al. 2007). The global environmental change tradition evolved from social science research in the 1970s concerning environmental hazards

(Blakie et al. 1994). This approach gained considerable momentum as adaptation became a central focus of the Intergovernmental Panel on Climate Change (Smit and Pilifosova 2001). The IPCC defines adaptation as “the adjustment in natural or human systems in response to actual or expected climatic stimuli or their effects, which moderates harm or exploits beneficial opportunities” (Parry et al. 2007, p. 6). Environmental change research on adaptation takes an actor-centered approach and focuses on how people respond to environmental stressors and what factors influence social actors’ response to stressors (Adger et al. 2005b, Smit and Wandel 2006, Nelson et al. 2007). The resilience approach to adaptation, on the other hand, emphasizes functioning of a system as a whole. For example, a society may be able to adapt well to a stressor from a social perspective (e.g. improved irrigation), but the ecological consequences of adaptation must be considered as well (e.g. ecological impacts of ground water pumping) (Nelson et al. 2007). Resilience thinkers argue that change is a fundamental aspect of any system and that adaptive dynamics are an inherent property of social-ecological systems (Berkes et al. 2003). Adaptation is seen as a function of the available resources, including financial, human, natural, physical and social capital, and the capacity of the system to respond to disturbances (Tompkins and Adger 2004). My research employs an adaptation approach to investigate fishers’ capacity to respond to change.

Vulnerability

Vulnerability is a related analytical approach for understanding the response of social-ecological systems to stress or perturbations (Miller et al. 2010). In contrast to resilience, which has its roots in ecology, vulnerability research has been influenced by theoretical traditions in hazards research, human ecology, political economy and political ecology (Eakin and Luers 2006). Vulnerability is most often defined as the degree to which a system is susceptible to and is unable to cope with stressors (Parry et al. 2007). Vulnerability researchers are concerned with how responses to system stressors are socially differentiated (Miller et al. 2010). It is a core concept in the study of livelihoods and poverty (Chambers and Conway 1992). Importantly, vulnerability is seen as being shaped by historical processes, differential entitlements and power relations, rather than as a direct outcome of perturbation or stress (Blakie et al. 1994, Eakin and Luers 2006). While there are differences in vulnerability research approaches, vulnerability is commonly characterized as being a function of exposure, sensitivity and adaptive capacity

(Adger 2006, Parry et al. 2007, Marshall et al. 2010). The central concern of both vulnerability and resilience analyses is how people develop the capacity to cope with and live with change.

Fishers’ Knowledge

In resilience thinking, a focus on uncertainty in social-ecological system dynamics has led to questions about the limits of scientific knowledge (Holling 1993). Resilience thinkers

emphasize the importance of recognizing unknowns within our systems of study. This openness to unknowability in resilience thinking has been crucial for creating legitimate space for the incorporation of other types of knowledge into scientific inquiry, such as fishers knowledge (Cote and Nightingale 2012). Fishers’ knowledge forms the core of the data for my dissertation. Following the seminal works of Bob Johannes (1981) and Firket Berkes (2012), I see fishers’ knowledge as critical for sustainable governance of fisheries. I use the term fishers’ knowledge to refer to “a cumulative body of knowledge, practice and belief, evolving by adaptive processes and handed down through generations by cultural transmission, about the relationship of living beings (including humans) with one another and with their environment” (Berkes 2012, p. 7). Often, local resource users are aware of changes in the resource base long before they have been detected by western science. Thus, fisher knowledge can complement scientific knowledge, not only in terms of adding to the range of information available, but also in terms of scale (Haggan et al. 2007, Berkes 2011). In Mozambique, fishers’ knowledge on historical changes in the coastal social-ecological system is particularly valuable since long-term data sets are unavailable (Johannes et al. 2000). Throughout the research, I compared fishers’ accounts with other sources of data (e.g. reports of drought with national precipitation data) and consistently found that they were aligned, thus validating both sources of knowledge. I have attempted to take a

conscientious approach to the research and remain cognizant of the complexities involved in working with fishers’ knowledge (Silver and Campbell 2005, Shackeroff and Campbell 2007).

Livelihoods

Based on the literature on capabilities (Sen 1981), assets (Swift 1989) and sustainability (WCED 1987), Conway and Chambers (1992) proposed the sustainable livelihoods approach as a means for understanding the lives of poor people. According to Ellis (2000, p. 10), “a livelihood

to these (mediated by instititions and social relations) that together determine the living gained by the individual or household”. The sustainable livelihoods framework recognises that households may be vulnerable to trends, shocks, seasonality and other factors from multiple scales that affect livelihood sustainability (Scoones 1998). A livelihood is considered to be sustainable when it can cope with and recover from shocks and stresses, and maintain and enhance its capabilities and assets both now and in the future, while not undermining the natural resource base (Chambers and Conway 1992). Stresses are characterized as continuously or slowly increasing pressure (e.g. chronic poverty); whereas shocks are characterized as acute spikes in pressure beyond the normal range of variability, usually discrete in space and time (e.g. rapid outbreak of a disease) (Turner et al. 2003, Marschke and Berkes 2006). In the livelihoods literature, assets are divided into five categories: financial capital (savings, credit), human capital (education, health), natural capital (land, trees, fish stocks), physical capital (infrastructure, material possessions), and social capital (kinship networks, associations) (Allison and Ellis 2001). Of interest is how people draw on different patterns of assets and institutional structures to adapt to stresses or shocks (Chambers and Conway 1992). Understanding how livelihood adaptation is negotiated at the household level is critical for strengthening existing adaptive capacity in small-scale fishing communities, for improving fisheries management and for rebuilding threatened communities (Allison and Ellis 2001, Coulthard 2008, Khan and Neis 2010).

Resilience, adaptation, vulnerability, and livelihoods frameworks are therefore appealing, and complimentary, perspectives because they offer dynamic and forward looking approaches for studying social-ecological change (Cote and Nightingale 2012). Together, these perspective can provide insights into the conditions under which some people, places or groups may be harmed by a given stressor, while others emerge either unharmed or in an improved position to withstand future shocks and stressors.

Introducing the Case Study

Why Mozambique?

Mozambique provides a particularly fascinating case for studying fishers’ responses to social-ecological change for a number of reasons. First, fishing has been critical to Mozambican

livelihoods for centuries. In the early 1500s, when Portuguese explorers first arrived in Mozambique an estimated 10,000 people were fishing along the central coast (Ehnmark and Wastberg 1963). Today, the fisheries sector constitutes 28% of national export earnings (Menezes et al. 2011). Moreover, in 2007 over 370,000 fishers and collectors were engaged in the small-scale fishery (IDPPE 2009).

Second, very little information exists on the small-scale fishery in Mozambique (Johnsen 1992). Limited government resources, civil conflict until 1992 and the use of Portuguese as the official language have discouraged scholarly research in the country (Pitcher 2002). As a result,

landings data for the small-scale fishery are only available for the last ten years (FAO 2011). In addition, only a handful of social science studies have been conducted on the fishery since independence (Menezes et al. 2009, 2011). This data-poor environment is troubling when we consider that the small-scale fishery accounts for more than 80% of total national landings and forms the basis of livelihoods for hundreds of thousands of people along the coast (FAO 2011).

Third, overfishing threatens the sustainability of social-ecological coastal systems in

Mozambique. In response to the lack of data on the small-scale fishery, Jacquet et al. (2010) reconstructed estimates of historical landings in Mozambique. Their study indicates that landings for both the industrial and small-scale fleets peaked in the mid-1980s and have subsequently declined. The severity of this decline is made more apparent considering that Mozambique is one of the poorest countries in the world. In 2009, over half of the population was living below the national poverty line of 18 meticais of US$0.50 a day (GoM 2011). Four years later, Mozambique was ranked 185 out of 187 on the UNDP’s Human Development Index (UNDP 2013). Formal employment in the country is very limited. Consequently, the small-scale fishery provides one of very few reliable livelihoods options for millions of people along the coast.

Finally, Mozambique offers a compelling case because it is a country characterized by change. Over the last 50 years, the country has experienced four distinct periods of transition:

colonialism, periods of war (independence war 1964-1975; civil war 1977-1992), socialism and free-market capitalism. These significant changes occurred over a relatively short period of

time, making Mozambique an informative case for studying how rural communities negotiate livelihood under conditions of constant change.

The Small-Scale Fishery in Mozambique

Small-scale fishing forms the basis of livelihoods for hundreds of thousands of people in coastal Mozambique. In this dissertation, I use the term small-scale to mean vessels less than 10 meters in length, which is consistent with the Government of Mozambique’s definition of the small-scale fishery (Afonso 2006). The majority of vessels in Mozambique are non-motorized three to five meter dugout canoes or seven to nine meter flat bottom wooden boats called lanchas

(Johnsen 1992). Fishing is usually conducted in small groups (< 20) using seine nets, gill nets and hand lines. The small-scale fishery is a multi-species fishery focused primarily on shallow-water shrimp (e.g. Acetes erythraeus) and small pelagics such as sardines and anchovies (IIP 2005). Generally, a small portion of the catch is sold fresh for local markets; the rest is salted, smoked or dried and consumed by fishers and their households (Johnsen 1992, IDPPE 2009). The coastal fishery is treated as common property. National regulations include a minimum two inch mesh size on beach seines, to reduce the catch of juveniles, and a closed period from

December until March. Small-scale fishers’ livelihoods are diverse and adaptive. Most households involved in small-scale fisheries also depend on other livelihood activities such as farming, firewood collection, trade and transport (Menezes et al. 2011).

Research Approach

Place-Based, Case Study Analysis

I address the dynamics of livelihoods and resilience through placed-based, case study research. Placed-based, case study research is defined as descriptive analysis of a spatially linked,

distinctive combination of social and environmental conditions (Turner et al. 2003). The rational for analyzing local articulations of a larger phenomenon is that the case may be “unique but is not singular” (Castree 2005, p. 541). That is, case study research assumes that analysis of a case can generate data that are useful for the specificity of place and for building general theory (Turner et al. 2003). Further, Castree argues that case study research “shows the world to be persistently diverse. Yet it shows that this diversity arises out of multiscaled relations such that it does not emerge sui generis.” (2005, p. 541). Similarly, resilience research emphasizes the

importance of investigating the influence of processes operating at scales above and below the system of interest for understanding system dynamics (Gunderson and Holling 2002). By addressing the particularities of specific social-ecological systems, and being cognizant of multiscalar interactions, place-based, case study research can contribute to our understanding of the dynamics of livelihoods and the resilience of social-ecological systems. Just as studies of social-ecological resilience have helped to illuminate particular aspects of Mozambique’s experience, Mozambique offers a useful comparative case for the study of theoretical questions related to coastal communities’ response to change.

Mixed Methods

A complex social-ecological system cannot be captured using a single perspective (Berkes 2011). Rather, it is best understood through the use of hybrid approaches, spanning qualitative and quantitative traditions and built on stakeholder engagement (Miller et al. 2010). As Berkes et al. (2003, p. 7) argue:

Recognizing the importance of qualitative analysis is one consequences of the recognition of complex system phenomena for natural resource management. By qualitative analysis we mean the understanding of the system’s behavior to help guide management directions. Qualitative analysis follows from the nature of nonlinearity. Because there are many possible mathematical

solutions to a nonlinear model and no one ‘correct’ numerical answer, simple quantitative output solutions are not very helpful (Capra1996). This does not imply that quantitative analysis is not useful. Rather, it means that there is an appropriate role for both quantitative and qualitative analyses, which often complement each other.

By combining robust quantitative data with the rich narratives of qualitative assessments, mixed methods research more closely reflects the complexities and contradictions of real lives (Adger 2006). By employing a mixed-methods research approach, I aim to ask better questions, which target gaps in our understanding of social-ecological system dynamics.

Field Research

To evaluate small-scale fishers’ response to change in the Western Indian Ocean, I spent nine months in Mozambique between 2009 and 2012 (Table 1.1). In 2009, I conducted an initial field trip which allowed me to discuss project ideas with local stakeholders including fishers, village

heads, professors and students from the University of Eduardo Mondlane’s School of Marine and Coastal Sciences (UEM-ESCMC ) and employees from the Ministry of Fisheries Institute for the Development of Small-Scale Fisheries (IDPPE). With the assistance of local partners, I selected three research communities: one coastal fishing community (Zalala Beach), one estuarine fishing community (Inhangome) and one community engaged in export-oriented shrimp farming

(Inhanssunge). Communities were selected due to their similarity in size, close proximity to one another (~30 km separates the furthest two communities) and their dependence on three different marine-resource based livelihood strategies. During my first field trip, I received permission from the village head of each community to conduct my research. I conducted the majority of the data collection between September and December 2010. In 2012, I returned to Mozambique to collect some final data and verify initial results with the research communities.

Table 1.1 Timeline of fieldwork in Mozambique.

J F M A M J J A S O N D 2009

Initial field trip, site selection 2010

Primary data collection 2012

Data collection and verification

I used a combination of qualitative and quantitative research methods. Qualitative methods included interviews, participant observation, focus groups and literature reviews. Interviews were conducted in Portuguese, Chuabo or Sena with translation assistance from a Masters student from UEM-ESCMC and an IDPPE extension agent. Interviews typically lasted over an hour. In total, I conducted 86 interviews with fishers, 5 with fish traders (called compradores), 2 with IDPPE extension agents and 2 with village heads. In 2012, I conducted 2 focus groups, one at Zalala Beach and one in Inhangome, focusing on livelihood resilience. Details of the

interviews and focus groups are found in the respective dissertation chapters and in Appendices A and C-E. I reviewed the peer-reviewed and grey literature, such as dissertations, technical papers and project reports, which are an important source of information, particularly in Africa where peer-reviewed publications are often limited (Jiddawi and Öhman 2002). In addition, the time spent in Mozambique also allowed me to engage in many informal discussions with fishers, peers at UEM-ESCMC and the IDPPE, managers at the shrimp farm and friends which added

substantially to my understanding of the complexities of marine based livelihoods in central Mozambique.

Quantitative data included household surveys and fisheries statistics. I conducted 268 household surveys in three communities. Surveys were conducted in Portuguese, Chuabo or Sena with the assistance of students from UEM-ESCMC and typically lasted under 30 minutes. Specific details on the surveys are found in the respective dissertation chapters and in Appendix B. Landings data were retrieved from the FAO’s FishStat Plus database (2011). Data on the number of small-scale fishers in Mozambique were retrieved from the Government of Mozambique’s small-scale fishers’ census (1998, 2004, 2009).

Data analysis was based on triangulation of data from field notes, interviews, surveys and

external sources including the peer-reviewed and grey literature. Qualitative data were coded by hand or using qualitative software (NVivo 9). Descriptive statistics for quantitative data were calculated and graphed using Microsoft Excel 2010.

Structure of Dissertation

My dissertation is based on four manuscripts, each of which can stand on its own but contributes to a larger narrative that runs throughout. The order of the chapters is purposeful. Chapter 2 describes how the coastal fishery system changed in the past, Chapters 3 and 4 describe how fishers are currently responding to multiple sources of change and Chapter 5 decribes what the small-scale fishery system might look like in the future.

In Chapter 2, Historical perspective and recent trends in the coastal Mozambican fishery, I combine national landings statistics with career history interviews with fishers to generate a multi-scale historical reconstruction that describes social-ecological interactions over time. The chapter is based on the idea that the behaviour of a system depends not only on its current structure, but also on its past. At the national level, our analysis points towards trends of fishing intensification and decline in targeted stocks. At the local level, fishers describe a range of drivers of and responses to change. We argue that historical reconstructions based on data of

multiple types and scales will capture the complexities that characterize interactions between fish, fishers and their social-ecological context more effectively than any single data type.

In Chapter 3, Strengthening threatened communities through adaptation: insights from coastal

Mozambique, I investigate how fishers from two coastal communities are currently adapting to

livelihood stress, by assessing household assets, adaptive strategies and factors that facilitate or inhibit adaptation. I find that adaptations are spatially and social differentiated and place-specific. My data suggest that it is not the poorest fishers who are least able to adapt to change, but fishers who are locked into a declining fishery.

In Chapter 4, Impacts of export-oriented shrimp aquaculture on livelihood vulnerability: a case

from the Western Indian Ocean, I analyze how employment at a shrimp farm influences

household vulnerability. Through the Turner et al. (2003) vulnerability framework, I evaluate exposure, sensitivity and adaptive capacity of shrimp farm employees and non-farm employees from a rural community in central Mozambique. I find that employment at the shrimp farm reduces employees’ vulnerability to chronic stressors, but increases vulnerability to acute shocks associated with shrimp production.

In Chapter 5, Assessing fishing communities’ sustainability: insights from resilience thinking, I apply the Cumming et al. (2005) resilience framework to two fishing communities. By having fishers’ describe how they would respond to hypothetical decreases in catch rates, I identify social thresholds. After a 90% decline in catch rates, both communities would shift into qualitatively different systems yet the states would be different. Fishers from Zalala Beach would permanently relocate to find higher catch rates, while fishers from Inhangome would change professions. I discuss the implications for social-ecological sustainability and resilience theory.

In Chapter 6, I conclude my dissertation by summarizing seven key research findings of my doctoral research. Finally, I discuss some of the theoretical, methodological and policy

2. Historical Perspectives and Recent Trends in the Coastal

Mozambican Fishery

Blythe, J.L., G. Murray and M. Flaherty. 2013. Historical perspectives and recent trends in the coastal Mozambican fishery. Ecology and Society (in press).

Abstract

Historical data describing changing social-ecological interactions in marine systems can help guide small-scale fisheries management efforts. However, while fisheries landings data are often the primary source for historical reconstructions of fisheries, we argue that reliance on data of a single type and/or from a single-scale can lead to potentially misleading conclusions. For example, a narrow focus on aggregate landings statistics can mask processes and trends

occurring at local scales, as well as the complex ‘social’ changes that result from and precipitate marine ecosystem change. Moreover, in the case of many small-scale fisheries, landings statistics are often incomplete and/or inaccurate. In this paper, we draw on case study research in

Mozambique that combines national landings statistics with career history interviews with fish harvesters to generate a multi-scale historical reconstruction that describes social-ecological interactions within the coastal Mozambican fishery. At the national level, my analysis points towards trends of fishing intensification and decline in targeted species, and highlights the significant impact of small-scale fisheries on marine stocks. At the local level, fishers are experiencing changes in fish abundance and distribution, as well as in their physical, social and cultural environments and have responded by increasing their fishing effort. We conclude with a discussion of the governance implications of my methodological approach and findings.

Introduction

Few people would claim to know as much about how to catch fish as a good full-time fisherman. When it comes to understanding fish behavior and the many environmental factors that help determine and predict it, marine biologists must often take a back seat. This is hardly surprising. There are hundreds of times as many fishermen today as there are marine biologists, and their forebears were plying their trade and passing on their accumulated knowledge tens of centuries before anyone heard of marine biology. What

is surprising is how little effort has been made by scientists to search out and record this information.

- Johannes (1981, p. vii)

Coastal ecosystems around the world are being restructured by overfishing and other factors (Jackson et al. 2001, Myers and Worm 2003). In response, fisheries managers and researchers are calling for more nuanced understandings of the complex historical interactions between fishers and marine systems (Johannes 1981, Murray et al. 2006; St. Martin et al. 2007).

Developing effective fisheries governance and management responses requires an understanding of social–ecological transformations, including major trends and the suite of behaviors,

rationales and motivations that drive the interactions between fishers and their environments at different scales (Ludwig 1993, Ommer 2007).

Landings data are often the primary source for historical reconstructions of fisheries (Pauly et al. 1998, Garcia and de Leiva Moreno 2003). When combined with macro-level historical

information and analysis, they can be useful for interpreting macro-level social-ecological interactions. However, we argue that reliance on data of a single type and/or from a single-scale can lead to misleading conclusions. For example, a narrow focus on aggregate landings statistics can mask processes and trends occurring at local scales, as well as many of the complex social changes that both result from and create marine ecosystem change (Murray et al. 2007).

Furthermore, in developing countries, landings data are often incomplete and underestimate total catch, particularly by the small-scale sector (Zeller et al. 2007, FAO 2011). The

under-representation of small-scale fisheries in national landings is due to a variety of factors, including political marginalization linked to their relatively low contribution to gross domestic product and the cost associated with monitoring spatially diverse, and often physically remote, fisheries by resource-limited governments (Béné 2003, Zeller et al. 2006, Jacquet et al. 2010). In

Mozambique, for example, national catch data for the small-scale sector have been calculated using a sample of provinces which excluded the province with the largest number of active boats

(Jacquet et al. 2010). In these cases, there is a need to supplement aggregate landings statistics, such as those supplied to the Food and Agricultural Organization (FAO), with other data in order to better understand the dynamics of intensification, expansion, and stock depletion in coastal small-scale fisheries (Pauly and Zeller 2003; Murray et al., 2006; Murray et al., 2008; Berkes, 2003).

Fishers often possess a profound understanding of complex marine systems based on long-term interactions with their environment (Neis et al. 1999, Berkes et al. 2000). This accumulated knowledge, passed down through generations, is an important source of information about historical changes in local marine resources as well as social changes in marine social-ecological conditions (Johannes et al. 2000). In addition, fishers can provide critical information –

particularly in data poor situations – on changes in stock distribution and abundance as well as changes in fishing effort and fishing practices that are critical for interpreting catch-rate data (Haggan et al. 2007).

In this paper, we draw on case study research in coastal Mozambique that combines national landings with career-history interviews with fish harvesters to generate a multi-scale historical reconstruction that describes social-ecological interactions within the coastal fishery, and provides insights into the utility of combining these types of data. We begin by presenting our methodology. The results and discussion section is presented in two sections: macro-scale restructuring based on landings data (1950 – 2009) and micro-scale restructuring based on fishers’ knowledge. We conclude with a discussion of the major trends and the implications of a multi-source data approach for fisheries governance.

Methods

Study Site

In 2009, with the assistance of local researchers from the University of Eduardo Mondlane's School of Coastal and Marine Science (UEM-ESCMC) and staff at the Institute for the Development of Small Scale Fisheries (IDPPE), we selected Zalala Beach as our study site, because of its location near the productive Sofala Bank and its long history of both small-scale and industrial fishing. Zalala is located thirty kilometers north of Quelimane, on Mozambique’s

central coast (Figure 1). Zalala has a multi-species fishery focused primarily on shallow-water shrimp (e.g. Acetes erythraeus) and pelagics such as sardines and anchovies (IIP 2005). Fishers fish from dugout canoes or larger wooden ‘lancha’ vessels that the government of Mozambique defines as small-scale (Afonso 2006). Lanchas can accommodate crews of twenty fishers and are typically rowed or sailed. In 2009, there were two vessels with motors at Zalala Beach (A. Camunada, IDPPE, pers. comm.). Fishers primarily use gill nets, seine nets and hand lines (IDPPE 2009). Catches landed at the beach are dried (24.6%), salted (22.3%), smoked (23.4%), frozen (14.2%), or sold fresh (15.5%) (IDPPE 2009). The majority of fishers at Zalala Beach speak Chuabo or Sena as their first language and Portuguese as their second. Access to central markets in Zalala is limited by poor roads and limited motorized vehicle transportation. The majority of fish are sold to markets in surrounding districts by middlemen (called

‘compradores’) or consumed by fishing households.

Figure 2.1 Map of Mozambique (grey). Zalala Beach, our study site, is located 30 kilometers north of Quelimane.

Macro-Scale Data

Landings data were retrieved from the FAO’s FishStat Plus database (2011). They consist of total national marine capture production in Mozambique’s exclusive economic zone between 1950 and 2009. Data include all quantities of fish, crustaceans and molluscs caught and landed, but exclude discards, harvest from aquaculture and marine mammals, inland capture, sponges and aquatic plants. Data on the number of small-scale fishers in Mozambique were retrieved from the IDPPE small-scale fishers’ census (1998, 2004, 2009). Landings and fishers data were imported into, and graphed in, Microsoft Excel 2010. In order to describe socio-political, economic and governance trends at the macro level over the last five decades, we consulted the peer-reviewed literature on fisheries in Mozambique and publications from the Ministry of Fisheries. We also reviewed the grey literature, such as dissertations, technical papers and project reports, which are an important source of information, particularly in Africa where

peer-reviewed publications are often limited (Jiddawi and Öhman 2002).

Interviews

In order to supplement landings and other macro-level data, we conducted semi-structured interviews, adapted for the study of fishing communities, with experienced fishers in Mozambique (McGoodwin 2001). Initial respondents were identified by the village head (‘Secretaria do Bairro’) based on two criteria: 1) that they derived their livelihood primarily from fishing and 2) that they had a minimum of ten years of fishing experience. Subsequent respondents were identified through snowball sampling by asking interviewees to suggest other experienced fishers at Zalala Beach (Davis and Wagner 2003). Interview questions were tested in a pilot study with students from UEM-ESCMC. We interviewed 15 fishers between

September and December 2010 and 10 fishers between May and July 2012, which coincided deliberately with the high and low fishing seasons respectively. The majority of interviews were conducted mid-morning, on the beach, as fishers returned from fishing. Interviews were

conducted in Portuguese, Chuabo or Sena with translation assistance from a Masters student from UEM-ESCMC and an IDPPE extension agent. Fishers were asked to describe changes in fish abundance and distribution, physical environment, social and cultural context, regulatory changes and fishing practices. Interviews typically lasted between 30 and 45 minutes.

attempt to understand fishers’ perceptions of and responses to change. For example, when discussing changing boat type, we focused on when this change occurred, why it occurred and how it impacted their catch. Interviews were coded using qualitative software (NVivo 9). Descriptive statistics (mean age, years fishing and daily catch) were calculated using Microsoft Excel 2010.

Results and Discussion

Macro-Scale Restructuring in the Mozambican Fisheries: 1950-2009

The last fifty years of Mozambique’s history can be divided into four fairly distinct socio-economic periods: colonialism, periods of war, socialism and free-market economy. Here we relate the available national landings data to these four periods (Figure 2.2).

Figure 2.2 a) Periods of socio-economic change affecting coastal fisheries (adapted from Menezes et al. 2011), and b) total landings, 1950-2009 (FAO 2011) and total number of small-scale fishers (IDPPE 1998, 2003, 2009) in Mozambique.

In the 16th century, when the Portuguese arrived in Mozambique, an estimated 10,000 people were living along the Sofala bank and engaging in fishing (Ehnmark and Wastberg 1963). For the majority of the colonial period, fishing remained largely subsistence based, as trawling was prohibited by colonial fishery law (Lopes and Gervasio 2003). In the early 1960’s, however, the Portuguese began to recognize the export earning potential of a shrimp fishery (Jacquet and Zeller 2007). They overturned the trawling ban, established a small industrial fleet that was owned and operated by crews from Portugal, and built large processing and freezer plants along the coast (Krantz et al. 1986, Menezes et al. 2011). During the colonial period, landings were not collected for the small-scale sector. However, it has been estimated that more than 16,000 rural fishers were active along the coast (Herrick et al. 1969). From the 1950s until the mid-1970s, national landings gradually rose (Figure 2.2).

Approximately 500 years of colonial rule came to a close when the Front for the Liberation of Mozambique (FRELIMO) initiated an armed campaign for independence in 1964. After a decade of conflict, which destroyed much of the country’s infrastructure, FRELIMO established an independent socialist state. The new government nationalized and invested in the industrial fishery following the colonial war (Menezes et al. 2011). In 1976, Mozambique established a 200 mile exclusive economic zone. In need of foreign exchange, the government formed joint enterprises with private fishing interests in Japan, Spain and Norway, and traded fishing rights for aid from the Soviet Union (Jacquet and Zeller 2007). Although Mozambique’s fishing grounds had not been fully surveyed, Norway suggested increasing the annual production of fish by 20,000 tonnes by 1985 through the adoption of bottom trawling (Avezedo 2002). By the early 1980s, shrimp made up over a quarter of total national landings and became Mozambique’s second largest earner of foreign exchange, cashews being first (Jacquet and Zeller 2007, FAO 2011). Supported by these investments, industrial landings grew and peaked in the mid-1980s (Figure 2.2). The new socialist government also invested in the small-scale fishery. They introduced fishing cooperatives (called ‘combinados pesqueiros’) in the early 1980s (Menezes 2009). The cooperatives focused on meeting state production targets and supported the small-scale fishers by providing fishing gear, building processing facilities and increasing access to central markets. This resulted in a new level of livelihood security in the small-scale sector (Menezes et al. 2011).

In 1977, a civil war erupted between the ruling socialist FRELIMO and the anti-communist Mozambique National Resistance (RENAMO), which led to mass migrations. Close to two million Mozambicans fled abroad, while approximately four million people were displaced internally (Azevedo 2002). The majority of internally-displaced people fled from fighting in the interior to coastal areas (Menezes 2009). Once along the coast, many refugees turned to fishing (Menezes 2008). Historical evidence suggests that the additional fishing pressure resulting from this migration led to declining landings during the civil war (Figure 2.2, Lopes and Gervasio 1999, Menezes 2011). During the civil war, little fisheries legislation was adopted and resource monitoring was limited (Afonso 2006). In 1992, both parties signed a cease fire agreement.

Towards the end of the civil war, the government began to transform the nation’s centrally planned economy into a free-market system. In 1987, Mozambique adopted the International Monetary Fund’s structural adjustment programs. In November 1990, the Constitution and first Fisheries Act were established (MoF 1995). These economic and political shifts created

significant reforms in the government’s relationship with small-scale fisheries. For example, fishing cooperatives were abolished (Menezes 2008), government service switched from direct intervention through the provision of services towards the creation of local governance

institutions (Menezes 2009), and the government began promoting fisheries co-management (MoF 1995). In addition, free-market reforms led to hardship in many fishing communities; schools and health centers were closed and direct government assistance for small-scale fishers was reduced (Menezes 2009, Menezes et al. 2011). Landings fluctuated through the 1990s, but never returned to their 1980s levels (Figure 2.2). Yet, since the end of the civil war Mozambique has experienced steady economic growth. The government has begun to rebuild its resource-management capacity and allocated resources to improve monitoring of the small-scale fisheries sector (Afonso 2006).

Clearly something unusual happened in 2003 (Figure 2.2). In that year, the government began reporting small-scale catch data to the FAO for the first time and landings jumped from 20,515 tonnes in 2002 to 79,451 tonnes in 2003 (FAO 2011). These data indicate that small-scale fishers are catching more than 75 per cent of total national landings, which is very significant

considering that data for the sector is only available since 2003. Fortunately, by combining estimates of individual catch rates and total number of small-scale fishers Jacquet et al. (2010) reconstructed total marine catch for Mozambique between 1950 and 2005. While there is a level of uncertainty associated with their estimates, their reconstruction present two important points. First, small-scale landings reported since 2003 were based on a sample of fishing centers and provinces, which were not extrapolated countrywide. Therefore, Jacquet et al. (2010) estimate that Mozambique’s annual catch rates were potentially between 47,000 and 177,000 tonnes higher than the reported data suggest. Second, reconstructed marine catch data indicate that small-scale landings peaked in the mid-1980s followed by a subsequent decline, which suggests overfishing of local resources (Jacquet et al. 2010). Thus, data provided to the FAO suggest that small-scale landings in Mozambique account for three-quarters of total marine catch and that they have been significantly under-reported. Reconstructed catch data suggest that national landings have declined since the 1980s, yet the explanatory power of the reconstruction data is limited by minimal input data and a high level of uncertainty associated with estimates. Alone, the existing data for the coastal Mozambican fishery cannot address the diversity of changes in fish abundance or distribution, location specific interactions of humans with physical

environmental heterogeneity, social, cultural or regulatory changes or the impact of new

technologies on marine landings. It is, therefore, critical to identify other sources of information that can complement the data provided to the FAO and contribute to our understanding of the interactions between small-scale fishers and coastal marine resources.

Micro-scale restructuring: fishers’ knowledge

During the last few decades, fishers along Mozambique’s coast have witnessed, participated in and responded to radical transformation within their social-ecological system. In this section, we present fishers’ descriptions of change in the fishery along the Sofala Bank (Table 2.1).